Hacker News Reader

Comment by jmspring | Hacker News Reader

Lost me at "based on Cygwin".

Yes Cygwin brings some unix goodness, but it has enough warts that I avoid it.

I wish I knew why so many Windows devs hate Cygwin. It has some pretty poor default settings, but you can change your .bashrc just like on Linux, or ditch the whole thing and use zsh instead, which works out of the box with oh-my-zsh (it looks like Babun does this as well.) For me, Cygwin is a very comfortable environment for git, programming and basic file management.

It's also the only way to get Linux shells to run on Windows, since they all rely on fork(). Even MSYS is just a fork of Cygwin 1.5.

I don't hate Cygwin, but anything using it should bill itself as "blah on top of Windows", not "blah for Windows".

(Or alongside if you want to say it that way.)

Yep. Give me something built on msys anyday.

Can you give some examples?

http://cygwin.com/faq/faq.html#faq.using.bloda

(not really Cygwin's "fault" per se, but it is what it is)

I almost got a 0 on a programming project for an intro to CS class because it compiled and ran perfectly fine on Windows, but segfaulted on Linux.

Which is much less a Cygwin problem than a poor understanding of the tools problem.

How is it not a cygwin (well, technically gcc for cygwin) problem? The port of gcc was bad, that's all there is to it. When I found a bug where a certain version of clang on ARM wouldn't compile a program everything else was compiling, was it a problem with my understanding or was it a bug with clang?

cygwin's ports aren't going to be 100% perfect, not sure why you would claim that if there's a problem with it, it's MY misunderstanding that's the problem.

Reply on news.ycombinator.com